BEGAN in PyTorch
PyTorch implementation of BEGAN: Boundary Equilibrium Generative Adversarial Networks.
Requirements
- Python 2.7
- Pillow
- tqdm
- PyTorch
- torch-vision
- requests (Only used for downloading CelebA dataset)
- TensorFlow (Only used TensorBoard for logging)
Usage
First download CelebA datasets with:
$ apt-get install p7zip-full # ubuntu
$ brew install p7zip # Mac
$ python download.py
or you can use your own dataset by placing images like:
data
└── YOUR_DATASET_NAME
├── xxx.jpg (name doesn't matter)
├── yyy.jpg
└── ...
To train a model:
$ python main.py --dataset=CelebA --num_gpu=1
$ python main.py --dataset=YOUR_DATASET_NAME --num_gpu=4 --use_tensorboard=True
To test a model (use your load_path
):
$ python main.py --dataset=CelebA --load_path=logs/CelebA_0404_105537 --num_gpu=0 --is_train=False
Results
Failed to reproduce the result of the paper. Looks like k_t
doesn't head to gamma
and there is a mode collapsing problem (not sure how the paper deals with this without pulling-away term of EBGAN).
(in progress)
Author
Taehoon Kim / @carpedm20